Chronic Ketogenic Low Carbohydrate High Fat Diet Has Minimal Effects on Acid–Base Status in Elite Athletes

lthough short (up to 3 days) exposure to major shifts in macronutrient intake appears to alter acid–base status, the effects of sustained (>1 week) interventions in elite athletes has not been determined. Using a non-randomized, parallel design, we examined the effect of adaptations to 21 days of a ketogenic low carbohydrate high fat (LCHF) or periodized carbohydrate (PCHO) diet on pre- and post-exercise blood pH, and concentrations of bicarbonate [HCO3 ] and lactate [La ] in comparison to a high carbohydrate (HCHO) control. Twenty-four (17 male and 7 female) elite-level race walkers completed 21 days of either LCHF (n= 9), PCHO (n= 7), or HCHO (n= 8) under controlled diet and training conditions. At baseline and post-intervention, blood pH, blood [HCO3 ], and blood [La ] were measured before and after a graded exercise test. Net endogenous acid production (NEAP) over the previous 48–72 h was also calculated from monitored dietary intake. LCHF was not associated with signi cant differences in blood pH, [HCO3 ], or [La ], compared with the HCHO diet pre- or post-exercise, despite a signi cantly higher NEAP (mEq day 1 ) (95% CI = (10.44;

after a graded exercise test. Net endogenous acid production (NEAP) over the previous 48–72 h was also calculated from monitored dietary intake. LCHF was not associated with signi cant differences in blood pH, [HCO3 ], or [La ], compared with the HCHO diet pre- or post-exercise, despite a signi cantly higher NEAP (mEq day 1 ) (95% CI = (10.44; 36.04)). Our results indicate that chronic dietary interventions are unlikely to in uence acid–base status in elite athletes, which may be due to pre-existing training adaptations, such as an enhanced buffering capacity, or the actions of respiratory and renal pathways, which have a greater in uence on regulation of acid–base status than nutritional intake. Keywords:dietary interventions; periodized carbohydrate diet; fat adaptation; keto-adaptation 1. Introduction Low carbohydrate high fat (LCHF) diets have previously been implemented in the context of epilepsy treatment [1,2] and as a weight loss strategy [3,4]. More recently, however, there has been a re-emergence of interest in their potential role in sports nutrition, with claims that adaptation to restricted carbohydrate (CHO) intake and high levels of circulating ketone bodies by trained individuals achieves signi cant changes in substrate utilization during sub-maximal exercise, to shift reliance from glycogen utilization to the relatively unlimited stores of body fat [5]. Indeed, Nutrients2018,10, 236; doi:10.3390/nu10020236

Nutrients2018,10, 236 2 of 13 both early [6] and more recent [7–9] studies have shown that sustained (3 weeks to several years) exposure to such a diet causes these robust shifts in exercise fuel use. However, bene cial effects on performance remain unsubstantiated, with reports of maintained capacity for submaximal cycling under fasted conditions in well-trained cyclists [6], but a reduction in exercise economy and failure to improve 10,000 m race performance following a block of intensi ed training in elite race walkers [9], in comparison to a more traditional diet providing high CHO availability [10]. A third approach to nutrition support for endurance sport is the periodized CHO diet, which integrates strategies to achieve high CHO availability to support key training sessions with protocols for low CHO availability to enhance adaptive responses to selected lower intensity sessions [11]. This redistribution of CHO intake to target the individualized goals of each training session has been shown to alter substrate utilization during submaximal exercise [12], and to produce performance bene ts in sub-elite [11,13], but not elite [9,14] athletes. Thus, it appears that sports performance is determined by factors other than a simple change in substrate utilization. One of the less well-studied effects of sustained manipulations of the macronutrient composition of the diet is the alteration in acid–base status. Diets high in fat can increase blood acidity, attributed to the stimulation of lipolysis, and the release of acidic ketone bodies [4]. A small number of studies have investigated the effects of short-term dietary modi cations on acid–base status and exercise capacity in healthy but essentially untrained individuals [15–18]. For example, measurement of pre-exercise blood pH and bicarbonate concentrations after an overnight fast showed that three days of a low (<10% energy intake) CHO diet was associated with a reduction in blood alkalinity, compared with a high (>65% energy) CHO diet [15]. Other studies by the same group con rmed the effect of various macronutrient concentrations. After a three-day high-protein (24% total energy intake), high fat diet (73% total energy intake) [17], and a three-day low-carbohydrate diet (4% total energy intake) [16],

intake) CHO diet was associated with a reduction in blood alkalinity, compared with a high (>65% energy) CHO diet [15]. Other studies by the same group con rmed the effect of various macronutrient concentrations. After a three-day high-protein (24% total energy intake), high fat diet (73% total energy intake) [17], and a three-day low-carbohydrate diet (4% total energy intake) [16], pre- and post-exercise blood pH and blood bicarbonate concentrations decreased, compared with a high carbohydrate diet. While these results indicate modi cations of acid–base status after acute alterations in dietary macronutrient intake, and potentially a separate and additional contribution to changes in exercise capacity, the effects have yet to be determined for sustained dietary interventions, particularly in athletic populations. Acid–base status has primarily been investigated in terms of the deleterious health implications of metabolic acidosis [19–22]. Damage to bone and muscle can occur when additional calcium is excreted in response to the excess of protons, and adverse effects, such as kidney stones, can be caused by a reduced urine pH, a further compensatory mechanism against acidosis [23,24]. Small changes to acid–base status place substantial stress on the body's buffering mechanisms [23,25], and one contributing factor is the composition of the diet [26]. It is acknowledged that normal metabolism incorporates many reactions that produce and consume acids and bases, and that acid–base status is routinely corrected by a tightly controlled acid–base regulatory system. The discrepancy between the acid and base forming reactions due to dietary intake is, however, the primary contributing factor toward net endogenous acid production (NEAP) [25]. NEAP can be calculated via validated equations [26,27]. One widely used method focuses on the consumption of protein and potassium, quanti ed over a 24 h period [26]; since protein is a sulfuric acid precursor, and potassium ingestion results in bicarbonate formation, the two dietary components have a substantial in uence on acid production [26]. Another more comprehensive estimation of acid–base status involves more variables, including the intake of protein and speci ed micronutrients (potassium, phosphate, magnesium, and calcium) over 24 h, body surface area (which affects acid excretion rates), and established

sulfuric acid precursor, and potassium ingestion results in bicarbonate formation, the two dietary components have a substantial in uence on acid production [26]. Another more comprehensive estimation of acid–base status involves more variables, including the intake of protein and speci ed micronutrients (potassium, phosphate, magnesium, and calcium) over 24 h, body surface area (which affects acid excretion rates), and established intestinal absorption rates of different dietary components [27]. Several previous studies have estimated acid production due to short-term (7 days or less) dietary interventions [15–18,28,29] using a similar principle to the two approaches described, but NEAP has not been calculated for sustained dietary interventions. NEAP calculations could potentially provide an indication of the effect of different dietary regimes on acid–base status, particularly when combined with direct measures of acid–base status, such as blood pH. There are potential implications for athletes'

Nutrients2018,10, 236 3 of 13 performance, given that an increased alkalosis or buffering capacity can improve performance, and a reduction in blood or muscle pH can be detrimental to high-intensity exercise performance [30,31]. Furthermore, differences in acid–base status between athletes and untrained individuals have been reported, associated with differences in buffering capacity [32,33]. The goal of the current study was to ll gaps in our current knowledge on the longer-term changes in acid–base status associated with diets that have been shown to alter blood pH and bicarbonate concentrations when followed for short periods. In particular, we wanted to investigate the effects of adaptation to a sustained ketogenic diet or periodized carbohydrate availability on such markers before and after exercise in elite athletes, and whether NEAP calculations of endogenous acid production were able to predict or explain any changes. Therefore, our aims were to determine the effect of adaptations to two sustained dietary interventions of current interest to endurance athletes (a LCHF diet and a periodized carbohydrate availability diet [9]) on blood bicarbonate concentration, blood pH, and blood lactate concentration (pre and post exercise), and net endogenous acid production (NEAP). 2. Materials and Methods A non-randomized parallel groups study design was used to determine the effect of three sustained (3 week) dietary interventions (a control high CHO diet—HCHO; low carbohydrate high fat diet—LCHF, and periodized carbohydrate availability diet—PCHO) on acid base balance, at rest, and following a graded exercise test in national and international level race walkers (Figure). Acid–base status was measured via capillary blood samples, which were analyzed for blood pH, blood bicarbonate [HCO3 ] concentration, and blood lactate [La ] concentration. Furthermore, calculations were undertaken to determine net endogenous acid production (NEAP) of the diets from dietary records kept over the 48 h prior to each exercise test. These measurements were conducted at two timepoints: baseline (prior to any intervention and with participants following a self-chosen diet), and post-testing (after a three-week dietary and training intervention, during which all food was provided to participants and consumed under supervision). The study was part of a larger research project, conducted during

over the 48 h prior to each exercise test. These measurements were conducted at two timepoints: baseline (prior to any intervention and with participants following a self-chosen diet), and post-testing (after a three-week dietary and training intervention, during which all food was provided to participants and consumed under supervision). The study was part of a larger research project, conducted during a residential training camp held throughout January and February 2017 at the Australian Institute of Sport (AIS; Canberra, Australian Capital Territory, Australia). All participants gave their written informed consent for inclusion, prior to their participation in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the AIS Ethics Committee (Project Approval Code: 20161201).Nutrients 2018, 10, 236 3 of 13 capacity can improve performance, and a reduction in blood or muscle pH can be detrimental to high‐intensity exercise performance [30,31]. Furthermore, differences in acid–base status between athletes and untrained individuals have been reported, associated with differences in buffering capacity [32,33]. The goal of the current study was to fill gaps in our current knowledge on the longer‐term changes in acid–base status associated with diets that have been shown to alter blood pH and bicarbonate concentrations when followed for short periods. In particular, we wanted to investigate the effects of adaptation to a sustained ketogenic diet or periodized carbohydrate availability on such markers before and after exercise in elite athletes, and whether NEAP calculations of endogenous acid production were able to predict or explain any changes. Therefore, our aims were to determine the effect of adaptations to two sustained dietary interventions of current interest to endurance athletes (a LCHF diet and a periodized carbohydrate availability diet [9]) on blood bicarbonate concentration, blood pH, and blood lactate concentration (pre and post exercise), and net endogenous acid production (NEAP). 2. Materials and Methods A non‐randomized parallel groups study design was used to determine the effect of three sustained (3 week) dietary interventions (a control high CHO diet—HCHO; low carbohydrate high fat diet—LCHF, and periodized carbohydrate availability diet—PCHO) on acid base balance, at rest, and

pH, and blood lactate concentration (pre and post exercise), and net endogenous acid production (NEAP). 2. Materials and Methods A non‐randomized parallel groups study design was used to determine the effect of three sustained (3 week) dietary interventions (a control high CHO diet—HCHO; low carbohydrate high fat diet—LCHF, and periodized carbohydrate availability diet—PCHO) on acid base balance, at rest, and following a graded exercise test in national and international level race walkers (Figure 1). Acid– base status was measured via capillary blood samples, which were analyzed for blood pH, blood bicarbonate (HCO 3 −) concentration, and blood lactate (La −) concentration. Furthermore, calculations were undertaken to determine net endogenous acid production (NEAP) of the diets from dietary records kept over the 48 h prior to each exercise test. These measurements were conducted at two timepoints: baseline (prior to any intervention and with participants following a self‐chosen diet), and post ‐testing (after a three‐week dietary and training intervention, during which all food was provided to participants and consumed under supervision). The study was part of a larger research project, conducted during a residential training camp held throughout January and February 2017 at the Australian Institute of Sport (AIS; Canberra, Australian Capital Territory, Australia). All participants gave their written informed consent for inclusion, prior to their participation in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the AIS Ethics Committee (Project Approval Code: 20161201). Figure 1. Overview of testing conducted in the study. All testing was conducted at baseline (prior to any intervention), and post‐testing (after a supervised three‐week training and dietary intervention). Figure 1. Overview of testing conducted in the study. All testing was conducted at baseline (prior to any intervention), and post-testing (after a supervised three-week training and dietary intervention).

Nutrients2018,10, 236 4 of 13 2.1. Participants Twenty-eight participants volunteered for this study, with twenty-four participants completing all testing (Table). Two males and one female were unable to complete the requirements due to injury, and one male was unable to provide a full data set for the baseline testing due to their delayed arrival at the training camp. The cohort consisted of elite race walkers, competitive at international or national level (mean SD International Association of Athletics Federations points = 1130 52) [34] who participated in a large dietary intervention study conducted at the Australian Institute of Sport with the support of Athletics Australia (further details are provided byBurke et al., submitted for publication).The study included participants from Australia, New Zealand, Poland, United Kingdom, South Africa, Lithuania, Canada, United States of America, Chile, Hungary, Japan, and Spain. A sample size estimation performed for the larger study, based upon outcome measures associated with performance and substrate utilization from recent research from our group with a similar cohort, indicated that eight participants per group would be required to detect physiological differences [9]. Participants were assigned to one of the two dietary intervention groups (LCHF or PCHO), or the HCHO control group, according to their preferred dietary intervention. Table 1. Participant characteristics (mean SD). Nutritional intake is presented as daily intake relative to body mass (BM), and as a percentage of total energy intake. Males (n= 18) Females ( n= 7) BASELINE TEST RESULTS Height (cm) 178.4 . 6.3 165.9 5.8 Mass (kg) 67.7 5.4 54.1 5.1 VO2max (mL kg min 1 ) 60.5 4.5 56.2 3.6 BASELINE NUTRITIONAL INTAKE Carbohydrate (g kg 1 BM) 6.6 1.3 (53 7%) 7.1 1.5 (53 9%) Protein (g kg 1 BM) 2.5 0.6 (20 4%) 2.5 0.6 (18 4%) Fat (g kg 1 BM) 1.5 0.5 (26 5%) 1.6 0.4 (26 5%) 2.2. Training Intervention Athletes followed a supervised and monitored training program across a three-week period of intensi ed training (Table). Of the ~12 weekly training sessions, six involved mandatory and supervised group sessions of race walking. There were two long sessions of 20 km,

0.6 (18 4%) Fat (g kg 1 BM) 1.5 0.5 (26 5%) 1.6 0.4 (26 5%) 2.2. Training Intervention Athletes followed a supervised and monitored training program across a three-week period of intensi ed training (Table). Of the ~12 weekly training sessions, six involved mandatory and supervised group sessions of race walking. There were two long sessions of 20 km, a track session involving sustained high-intensity 1-km repetitions on a 6-min cycle, a 14-km hill training session (change in elevation ~285 m), and two low–moderate intensity “recovery” sessions, while two resistance training and three hydrotherapy sessions were completed. The athletes could modify their remaining sessions, undertaking them as additional race walking sessions, or substituting with cross training, such as swimming or cycling. All athletes recorded their daily training in standardized training logs. Table 2. Template for weekly training program. Shading indicates mandatory sessions; remaining sessions could be modi ed by individual athletes. DAY Monday Tuesday Wednesday Thursday Friday Saturday Sunday AM 10 km walk 10–15 km walk # Resistance training >20 km long walk Hydrotherapy strategies 10 km walk Resistance training Hill session † >20 km long walk Hydrotherapy strategies 10 km walk or rest PM 1 km reps † Hydrotherapy strategies 10 km walk 10 km walk or rest 10–15 km walk 10–15 km walk # 10 km walk or rest # Indicates training session with low carbohydrate availability prior to the session (PCHO group only). † Indicates training session with low carbohydrate availability after the session (PCHO group only).

Nutrients2018,10, 236 5 of 13 2.3. Dietary Intervention All meals were prepared by chefs, according to standardized recipes, and were consumed by participants in a group environment, under the supervision of registered sports dietitians, according to a previously established practice [32]. For baseline testing, participants were able to choose freely from items provided as a buffet-style menu at the AIS Dining Hall, and were assisted to weigh and record all of their food intake during this period, using calibrated food scales (SJ-5001HS, A&D Weighing, Australia). Once the dietary intervention commenced, participants were provided with all meals and snacks according to their intervention group (PCHO or LCHF, or HCHO) with menus being individualized to BM and training load (to provide an energy availability of ~40 kcal kg 1 lean BM; LBM), and speci c dietary requirements including food allergies, intolerances, and dietary preferences. Each athlete could request an increase or decrease in the quantities of foods or drinks provided according to hunger, changes in training load, or uctuations in BM. Such variations were accommodated by adhering to the macronutrient composition of that individual's treatment group. During this period, all food intake was weighed prior to consumption, with allowances made for un nished portions or additional snacks chosen from a menu-speci c list, as per food diary logs. Energy and nutrient intake provided by the diets was calculated from a food analysis database speci c to Australian foods (Foodworks Version 9, Highgate Hill, Australia) by the same registered dietitian. Full details of the methodology for creating, providing, and recording food intake can be found elsewhere [35]. The three dietary interventions were chosen to represent different approaches to the support of high volume training programs [9]. - HCHO: traditional sports nutrition guidelines promoting high carbohydrate availability for all training sessions: CHO: ~8 g kg 1 BM, 60–65% energy intake; protein: ~1.8 g kg 1 day 1 15–20% energy; fat: ~20% energy intake [10]. - PCHO: contemporary approach to sports nutrition, with same energy and macronutrient composition as HCHO, but manipulated across and between days to provide high CHO availability for key training sessions

high carbohydrate availability for all training sessions: CHO: ~8 g kg 1 BM, 60–65% energy intake; protein: ~1.8 g kg 1 day 1 15–20% energy; fat: ~20% energy intake [10]. - PCHO: contemporary approach to sports nutrition, with same energy and macronutrient composition as HCHO, but manipulated across and between days to provide high CHO availability for key training sessions and low CHO availability for other sessions [11,36]. - LCHF: popular ketogenic low CHO high fat diet: CHO: <50 g day 1 , protein: ~1.8 g kg 1 day 1 15–20% protein, and 75–80% fat [6,34]. 2.4. Exercise Testing All participants completed a graded maximal exercise test at the baseline and post-testing timepoints, under controlled laboratory conditions. All participants conducted each test in a fasted state. The test was performed on a custom-built, motorized treadmill (Australian Institute of Sport, Canberra, Australia). The test was comprised of four submaximal stages (for determination of submaximal VO2and walking economy). Each stage, three minutes in length, was immediately followed by an incremental ramp to exhaustion for determination of VO2peak. As such, the total test duration was approximately 13 to 20 min, depending on a participant's time to exhaustion (TTE). The treadmill velocity for the rst stage was dependent on each participants' most recent 10 km race time (9–12 km h 1 ), at 0% gradient, with the velocity being increased by 1 km h 1 with each subsequent stage. After each stage, a small (5 microlitres; L) capillary blood sample was taken from the ngertip, to measure blood [La ] (Lactate Pro, Arkray, Kyoto, Japan). Immediately following the completion of the fourth submaximal stage (approximately equivalent to 20 km race-walk speed), the gradient of the treadmill was increased by 0.5 degrees every 30 s, until the participant reached volitional exhaustion. Heart rate (Polar heart rate monitor, Polar Electro, Kempele, Finland) was measured throughout the test. Expired ventilation samples were collected continuously throughout the test, using a custom-built open-circuit indirect calorimetry system (Australian Institute of Sport, Canberra, Australia). The system was calibrated prior to each test [37].

The study examines the effects of ketogenic and periodized carbohydrate diets on acid-base status in elite race walkers.